• No se han encontrado resultados

FASE 1: INVESTIGACION DOCUMENTAL

10. BIBLIOGRAFIA

Introduction

In order to analyze ZNF - RNA interactions, we have conducted preliminary studies to address secondary structural changes, binding localization on the RREIIBTR, orientation of binding for the protein, stoichiometry and specificity of binding. The various techniques mentioned in this chapter have been used in the following manner to illustrate interactions between RREIIBTR and zinc finger proteins.

1. Circular dichroism (CD) has been used to probe for any secondary structural changes that biomolecules (RNA or protein) undergo following the binding event. 2. Imino proton peaks indicate base pairing in nucleic acids. Detection of new peaks in the NMR imino proton spectra for the complex shows new base pairs formed, whose identity can be determined by NOESY cross peaks to previously identified unperturbed resonances in the free RNA. Additionally, chemical shift perturbation of the RNA imino peaks serves as a handy tool to localize the binding of the zinc finger proteins on the RNA.

3. In addition to determining the ZNF binding location on the RNA, it is also important to evaluate binding orientation. The !!" motif of zinc finger proteins gives them a cylinder-type geometry with the N-terminus and tip of the zinc finger representing one end of the cylinder while the C-terminus represents the other. We have substituted zinc in the ZNF with a paramagnetic metal ion with similar metal coordination geometry viz. Co2+. The metal coordination in the ZNF is situated closer to the C-terminus end of this “cylinder”. The broadening of specific

peaks in the imino proton spectrum of RREIIBTR, bound to this cobalt substituted ZNF, has allowed us to shed light on the binding orientation.

4. The changes in the ZNF protein backbone, on binding RREIIBTR, have been addressed by comparing the free and bound 15N HSQC spectra.

All the above-mentioned experimental techniques have also confirmed the stoichiometry and specificity of ZNFs binding to RREIIBTR.

Structural changes in RREIIBTR on ZNF binding: CD

CD spectra of biomolecules are used to analyze their secondary structures. We have compared the CD spectra of the free and ZNF29G29R bound RREIIBTR. Since the sensitivity of a CD signal is comparable to that of absorbance, the RREIIB CD signal dominates that of the protein in the complex CD spectrum.

All CD experiments were recorded at room temperature on a Jasco J-710 spectropolarimeter. The CD spectrum for the free ZNF29G29R was subtracted from that of the complex.

The free RREIIBTR CD spectrum is that of a typical right-handed A-RNA (1) (Figure 3.1) with a maximum near 260 nm, a minimum near 210 nm and 2 small negative peaks at 240 nm and between 290 and 300 nm. Upon addition of ZNF29G29R at a 1:1 stoichiometry, we see an slight increase in the intensity of the peaks at the above-mentioned wavelengths. These minor differences between the free and ZNF bound RNA suggests minimal changes if any, in the secondary structure of RREIIBTR, on ZNF binding. The marginal increase in

intensity might be a result of additional base pairing.

Figure 3.1 CD spectra of free RREIIBTR RNA (red) and ZNF29G29R complexed RNA (blue).

A 1:1 ZNF29G29R-RNA complex at a concentration of 10 µM in 10 mM sodium phosphate, 50 mM NaCl, pH 7.0, buffer at 298 K. Spectra (20 scans) were acquired in a 0.2 cm cuvette using a scanning speed of 200 nm/min.

Imino NMR

Labile protons in nucleic acids viz. imino protons, are visible in an NMR spectrum when their exchange with the solvent is slowed down, compared to the NMR time scale. This reduced exchange with the solvent is generally evidence of base pairing but could also indicate RNA–protein contacts in a complex or removal of a solvent accessible base to a solvent excluded region. Additionally, imino protons resonate further downfield than most other labile and non-labile protons of a biomolecule, thus providing a lesser degree of overlap and more amenable analysis. C D [ m d eg] Wavelenghth [nm]

Free RREIIBTR

The assignments of the imino resonances in the free RNA were obtained from previously published data (2) (Figure 3.2). All imino spectra henceforth had the same experimental parameters as stated in the legend of Figure 3.2, unless mentioned otherwise.

U43

A B

C

D

Figure 3.2 Imino NMR of free RREIIBTR.

A. Free RREIIBTR published assignments (2) B. Sequence of RREIIBTR (2) with regions highlighted: upper stem , middle stem , bulge , lower stem . C. Complete assignment of all RREIIBTR imino resonances. G41 and G55 were assigned from Figure 3.4. RNA sample conditions: 10 mM NaP buffer, pH 7.0, 100 mM NaCl, 200 µM !me, 298 K. NMR

spectra was collected using 1-1 jump and return pulse sequence (11). D. NOESY imino region of the free RREIIBTR with 150 ms mixing time and 1.5 s relaxation delay. The spectrum was collected using the NOESY version of the 1-1 jump and return pulse sequence. NOESY cross- peaks confirm assignments displayed in part C.

These assignments were also confirmed by the NOESY spectra shown in Figure 3.2 D and other truncated versions of RREIIBTR (Figure 3.4). The truncated versions of RREIIBTR also allowed us to assign G41 and G55, which were not published previously (2).

The assignments for the truncated versions of RREIIBTR were made by monitoring the sequential broadening of peaks in the imino spectra with increasing temperature. For example, a 12 base RNA oligonucleotide representing the upper stem of RREIIBTR was subjected to a temperature range of 283 – 303 K (Figure 3.3). The NMR imino spectrum has 5 resonances as expected, 4 base pairs and the solvent inaccessible G55. We observe that with increasing temperature, the peak at 12.66 ppm is the first to broaden followed by resonances at 13.98, 13.06, 13.43 and 10.75 ppm in succession. Though there are minor changes in chemical shift with increasing temperature, there is

Figure 3.3 Upper stem of RREIIBTR.

(left) Imino spectra of 12 base RNA oligonucleotide which represents the upper stem of RREIIBTR at temperatures 283, 293, 298, 303 K. (upper right) Sequence of the above mentioned RNA.

The numbering scheme is the same as used by (2).

5’

53

67

G55 G67 G64 G53 U66

66

adequate chemical shift dispersion to identify these peaks. Since the end of an oligonucleotide is the first to lose base pairing at increased temperatures, the resonance at 12.66 ppm is identified as G67. The other resonances were identified as they successively broadened with increasing temperature (Figure 3.3). It should be noted that G53 posseses higher thermal stability due to stacking and has been assigned accordingly.

We have used 2 truncated versions of RREIIBTR similarly (Figure 3.4), representing different sections of the free RNA, to identify and confirm the resonances presented in Figure 3.2.

RREIIBTR – ZNF comex 53 67 41 77 43

A

B

C

Figure 3.4 Assignment of RREIIBTR from its truncated versions.

A. Imino spectrum of RNA representing the lower stem of RREIIBTR (G41–C44 and G76-C79) with GCAA tetraloop. (RNA sequence on the right)

B. Imino spectrum of full length RREIIBTR along with its sequence (2) C. Imino spectrum of upper stem (C51 –G67) with RNA sequence.

RREIIBTR – ZNF complex

A titration of ZNF29G29R into RREIIBTR monitored by imino NMR spectra results in shifts of specific peaks and the appearance of new imino proton resonances, while some resonances do not show any shift at all (Figure 3.5). Both ZNF29 and ZNF29G29R had essentially superimposable imino spectra for their respective complexes with RREIIBTR. The resonances of the complex are broader than the free RNA, as would be expected due to the increased molecular mass for the complex. Additionally, there were no differences in the protein: RNA complex at 1:1 and 1.5:1 ratios (Figure 3.5). These observations indicate specific binding of both zinc finger proteins to RREIIBTR at 1:1 stoichiometry.

The 1:1 protein RNA complex showed distinct changes in chemical shift for U66 and U45. The NOESY spectra for the imino region identified only U45 due to its NOE cross-peak to G76 (Figure 3.7, I). No chemical shift changes were observed for imino protons near the upper stem (G53, G64) or near the 5’ – 3’ terminus of the lower stem (G42, U43, G76, G77) during the titration. This precludes these regions (Figure 3.5) from being at the binding location, thus restricting the zinc finger binding site to the same bulge region on the RNA that the Rev peptide utilizes.

The imino spectrum of the complex displayed new peaks at 13.3, 12.6, 12.65, 12.26 and 11.85 ppm other than the chemical shift changes for U66 and U45 (Figure 3.7, II). The peak at 12.84, which corresponds to G41 in the free protein, also appears intensified. The NOESY spectra did not show any cross peaks to these new peaks to aid in their identification. Also, unlike the NOESY

data for the RREIIBTR – Rev peptide complex (2), where a purine - purine base pair (G48 – G71, 12.2, 12.5 ppm) is formed on binding, we see no such intense cross peak that would indicate the same is occurring in the formation of our complex. In fact, a comparison of the RREIIBTR complex with the Rev peptide and that formed with our ZNF (ZNF29G29R) showed little similarity, except in the downfield shift of U66 (Figure 3.6).

In order to confirm that all new peaks in the complex belonged to the RNA, we have compared the 15N coupled and decoupled 1D NMR imino spectra of a

Figure 3.5 Titration of RREIIBTR with ZNF29G29R followed by NMR at 298 K.

Protein: RNA molar ratios 0:1, 0.5:1, 1:1, 1.5:1 were achieved by adding 20 µl aliquots of ZNF29G29R (1 mM stock) to a 100 µM RNA sample. The change in the chemical shift of the U66 resonance at 13.9 ppm is specific for RNA protein complex formation, have been indicated by dotted lines. The base pairs with unaffected imino proton chemical shifts after formation of a protein RNA complex are marked by gray boxes. Sequence of the RNA is displayed to the right with gray boxed regions corresponding to those shown in the imino spectrum. Note: This data has been published (13).

RREIIBTR (14N) – ZNF29G29R (15N labeled) complex. The 15N coupled and decoupled imino spectra are expected to show no differences if all peaks belong to the RNA. However, we observe splittings of ~ 80- 90 Hz in the 15N coupled

spectrum (Figure 3.7, II, A) at 13.3 and 12.26 ppm. This proves that these resonances originate from the protein. Subsequently, we identified these resonances to be His 23 # 2H and His 27 # 2H when we inspected the 10 -14 ppm region of a 1D 1H spectrum of the free ZNF in water (Figure 3.7, III). Thus

our complex has 3 visibly distinct new RNA imino peaks if we disregard the possibility of any new peak under the G41 peak and in the 12 -12.4 ppm region (Figure 3.7, II, B). The identity of these new RNA imino peaks has been complicated by the paucity of NOESY data to known resonances.

Figure 3.6 Comparison of RREIIBTR imino complex spectra with Rev and ZNF29G29R.

A. Published imino spectra of RREIIBTR-Rev peptide complex (2). B. Imino spectra of RREIIBTR-ZNF29G29R complex. Blue rhombuses indicate changes in chemical shift while red rhombuses indicate new imino peaks.

A

A

B

Free Protein

His 23 #2 H His 27 #2 H U45 - G76 G42 - G77 G46 - G77 U43 - G77 G76 - U43 G53 - G64 G42-U43 G77 U66 U45 G42 G53 G64 G76 G46 U43

(I)

(II)

(III)

Figure 3.7 Identification of new imino region resonances.

(I) NOESY spectrum of RREIIBTR-ZNF29G29R complex with assigned cross-peaks.

(II A) Imino spectra of RREIIBTR (14N) –

ZNF29G29R (15N labeled) complex with 15N

decoupler turned off. (II B) Imino spectra of RREIIBTR(14N) –

ZNF29G29R (15N labeled) complex with 15N

decoupler turned on. The red dotted lines are on the split peaks and the green dotted lines are the decoupled peaks. Only one of the split peaks is visible in both cases.The difference between the red and the green lines is approximately 45 Hz indicating a full splitting of ~ 90 Hz. These peaks belong to the protein and are marked by a black filled star while RNA peaks are marked by black filled circles). The carrier for the decoupler (15N) was set at 120 ppm.

(III) 1D 1H spectra of free ZNF29G29R in water in the same buffer and temperature conditions as II above acquired by 1-1 jump and return pulse sequence (11). The identity of the peaks were obtained from a 1D NOE difference experiment for the free protein.

Since there are no major rearrangements in the RNA secondary structure upon ZNF binding, as shown by CD, we hypothesized that the 2 new peaks (12.6 and 12.65 ppm) could originate from the base pairing of G50 - C69 and G70 - C49. These base pairs are not formed in the free RNA, but they are formed in the RREIIBTR-Rev complex (2). Additionally, it is possible that the broad peak at 11.85 ppm could originate from either stacking of the U72 base in the bulge or interaction of the U72 imino group with the protein. We also do not know if new peaks actually do arise under the G41 resonance or the overlapped region from 12 – 12.4 ppm. In order to address these hypotheses, we have designed a number of RNA mutants and variants and monitored their NMR imino proton spectra on ZNF binding (ZNF29G29R).

RNA variants were made by modifications in the bulge and the middle stem by replacing guanine bases by 2 - aminopurine. This base replaces a hydrogen for the 6-oxo moiety of G and thus does not have an imino proton. Hence, if any of the new peaks belong to the modified position, they would be absent in the modified RNA – ZNF complex spectrum. The RNA mutants were made by replacing the cytosine bases in the middle stem (above the bulge) by uracils. These RREIIBTR variants and mutants and their respective imino spectra on ZNF29G29R binding has been listed below in 8 datasets. References to upper stem, middle stem, bulge and lower stem are as indicated in Figure 3.2 B.

1. RREIIBTR_G50_2AP

In this RNA sequence the guanosine base at position 50 has been replaced by 2 - aminopurine. The free RNA imino spectrum (Figure 3.8 A) is

similar to that of free RREIIBTR albeit with slightly lower chemical shift dispersion in the 13–13.5 ppm region. In the free RREIIBTR_G50_2AP, the resonances in the vicinity of G50 (i.e., G67, U66) are broadened (Figure 3.8 A). The imino spectrum of the RREIIBTR_G50_2AP - ZNF29G29R complex is missing one of the new peaks at 12.65 (Figure 3.8B). This suggests that the peak at 12.65 ppm in the RREIIBTR-ZNF complex is from the G50 – C69 base pair.

2. RREIIBTR_G70_2AP

In this sequence, the guanosine base at position 70, above the bulge, is replaced by a 2 – aminopurine. We expect this modification to perturb the bulge and G46 has been assigned accordingly. In complex with ZNF29G29R we see a

Figure 3.8

Documento similar